Increased Prevalence of Thyroid Disease in Patients with Antineutrophil Cytoplasmic Antibodies–Associated Vasculitis

Increased Prevalence of Thyroid Disease in Patients with Antineutrophil Cytoplasmic Antibodies–associated Vasculitis

Maria Prendecki, Leyre Martin, Anisha Tanna, Marilina Antonelou, and Charles D. Pusey ABSTRACT. Objective. Antineutrophil cytoplasmic antibodies (ANCA)–associated vasculitis (AAV) has been linked with thyroid disease as a result of antithyroid medications. We assessed the prevalence of thyroid disease in our patients with AAV. Methods. Clinical records of 279 patients with AAV diagnosed between 1991 and 2014 were analyzed. Results. Thyroid disease was identified in 21.5% of patients, but only 2 had previously received propylthiouracil. There was a greater proportion of female patients, patients with antimyeloperoxidase antibodies, and patients with renal disease in the group with thyroid disease. Conclusion. Our data show a higher prevalence of thyroid disease in patients with AAV than the general population. This was not attributable to antithyroid drugs. (First Release March 15 2018; J Rheumatol 2018;45:686–9; doi:10.3899/jrheum.170661)

Key Indexing Terms: ANTINEUTROPHIL CYTOPLASMIC ANTIBODIES AUTOIMMUNE DISEASE VASCULITIS THYROID DISEASE

The etiology of autoimmune disease remains to be fully it was seemingly unrelated to antithyroid drugs. We therefore understood, and there are often interactions between genetic performed a retrospective analysis of patients with AAV in and environmental factors. There is a known association our center to identify the prevalence of thyroid disease and between different autoimmune diseases, both organ-specific its association with the use of antithyroid drugs. and systemic1. Autoimmune thyroid disease is documented to be associated with other organ-specific autoimmune MATERIALS AND METHODS diseases such as type 1 diabetes, Addison disease, and celiac We identified all patients with a diagnosis of AAV from a clinical database disease, and has also been associated with systemic of patients seen in our unit between 1991 and 2014. Patients were included regardless of different organ involvement; not all patients had evidence of autoimmune diseases such as systemic lupus erythematosus 2,3,4 renal involvement. Patients were excluded if they were diagnosed prior to and Sjögren syndrome . Antineutrophil cytoplasmic 1990 or if insufficient clinical information was available. A retrospective antibody (ANCA)–associated vasculitis (AAV) has been analysis of patient notes and laboratory data was carried out and data were reported to be associated with other systemic autoimmune collected on age, sex, ethnicity, ANCA specificity, organ involvement, and diseases such as antiglomerular basement membrane disease, evidence of renal impairment. We identified the presence of thyroid disease rheumatoid arthritis, and systemic sclerosis5,6,7,8. There have and use of antithyroid drugs. Because this was a retrospective study and all treatment decisions were been small case series and a case-control study associating 9,10 made prior to our assessment, research ethics approval was not required for thyroid disease and AAV . Some of these associate the this report, in accordance with the UK National Health Service Research induction of ANCA with antithyroid agents such as propyl- Ethics Committee guidelines. thiouracil (PTU) or carbimazole. However, other studies have Statistics. Mann-Whitney U test was used for continuous variables and shown either a low incidence of development of ANCA chi-square test for the difference in proportions between 2 groups. Logistic despite continued use of PTU, or absence of vasculitic regression was used for multivariate analysis and results are expressed as symptoms despite the development of ANCA11,12,13. In our OR with 95% CI. Results are reported as statistically significant when clinical practice, we noted a higher prevalence of thyroid p < 0.05. disease than would be expected in the general population and RESULTS From the Imperial College London, London, UK. Of the 325 patients with vasculitis who were managed in our M. Prendecki, MBBS, Imperial College London; L. Martin, MD, unit since 1991, 46 of these were excluded because of insuf- Imperial College London; A. Tanna, MBBS, Imperial College London; ficient information. Of 279 patients with AAV, 60 (21.5%) M. Antonelou, MBBS, Imperial College London; and C.D. Pusey, DSc, had evidence of thyroid disease: 49 (17.6%) patients had Imperial College London. hypothyroidism (5 patients with transient or subclinical Address correspondence to M. Prendecki, Department of Medicine, Imperial College London, Hammersmith Campus, Du Cane Road, London hypothyroidism), 10 had hyperthyroidism (5 patients had W12 0NN, UK. E-mail: [email protected] transient or subclinical hyperthyroidism), and 1 patient had Accepted for publication November 21, 2017. goiter without derangement of thyroid function. Of the

686 The Journal of Rheumatology 2018; 45:5; doi:10.3899/jrheum.170661

Downloaded on September 27, 2021 from www.jrheum.org patients with hyperthyroidism, 3 developed subsequent population14. The prevalence of hyperthyroidism was also hypothyroidism after antithyroid treatment. Two of the higher but the difference less marked, with 3.6% in our patients with derangement of thyroid function also had goiter population compared to a reported population prevalence of and 2 had thyroid nodules. There were 43 (71.7%) patients 0.5–2.0%14. This was seemingly independent of the use of who received treatment with thyroxine, 3 patients (5%) who antithyroid drugs, with only 2 documented cases of previous received radioiodine followed by thyroxine, and only 2 PTU use in our series. This is similar to a previously reported patients (3.3%) who were treated with PTU. prevalence of thyroid disease of 20% in 158 patients with The group of patients with AAV and thyroid disease was AAV, and 38% in women with AAV, in an American compared to those with AAV and no thyroid disease (Table case-control series; they also reported a low rate of use of 1). The patients with thyroid disease were more likely to be antithyroid drugs (2/129 patients)9. A Swedish study women than those without (73.3% vs 45.2%, p = 0.0002) and reporting comorbidities in patients with AAV found a slightly patients of Indo-Asian origin were also more likely to have lower prevalence of thyroid disease of 14.5%; in keeping thyroid disease than not (26.7% vs 14.2%, p = 0.02). A with our findings, they also reported that the increased preva- greater proportion of the patients with thyroid disease were lence is more striking in women, but there are no further data found to have evidence of renal disease (95.0% vs 81.7%, available regarding type of thyroid disease or treatment15. p = 0.02) as part of their vasculitis, and a smaller proportion Patients with AAV and thyroid disease were more likely had ENT involvement (25.0% vs 48.4%, p = 0.02; Figure 1). to have MPO-ANCA specificity than not. This association of More patients with AAV and thyroid disease were MPO vasculitis and thyroid disease has also been described antimyeloperoxidase (MPO) antibody–positive than negative in 1 small case series and in the case-control study of patients for MPO-ANCA (58.3% vs 34.7%, p = 0.0016). In multi- with AAV in the United States (all of whom had renal variate analysis, OR of patients with thyroid disease having disease)9,10. Most but not all our patients had evidence of MPO-ANCA specificity was 2.0 (p = 0.025), of being female renal involvement as part of their vasculitis, and thyroid was 3.3 (p = 0.0004), and of having renal organ involvement disease seemed to correlate with the presence of renal was 4.5 (p = 0.018; Table 2). disease; derangement of thyroid function has been associated with decreased kidney function in previous studies. The DISCUSSION mechanisms underlying this association are unclear, although In our cohort of 279 patients with AAV, the overall prevalence altered iodine handling in patients with low estimated of thyroid disease was 21.5%. The prevalence of hypothy- glomerular filtration rate and hemodynamic changes in roidism was 17.6%; this is much higher than the reported patients with thyroid disease have been suggested16,17. population prevalence of hypothyroidism in the United Thyroid disease also seemed to be less common in patients Kingdom, which is around 1%. This was particularly evident with ENT disease, although this is likely because of the in women for whom the prevalence of hypothyroidism in our inverse correlation of ENT disease with MPO-ANCA speci- cohort was 30.8% compared to around 2% in the general ficity. There were 58.3% of patients with proteinase-3 (PR3)– ANCA who had documented ENT disease, while only 20.7% Table 1. Characteristics of AAV patients with and without thyroid disease. of those had MPO-ANCA. There was no difference in Values are % unless otherwise specified. vasculitis outcomes between the patients with and without thyroid disease. Mortality rate and proportion of patients with Characteristics Thyroid No Thyroid Disease Disease stage 5 chronic kidney disease 5 at final followup were not History, History, significantly different between the 2 groups, although n = 60 n = 219 p detailed information on clinical outcomes was not collected. Human thyroid peroxidase (TPO) antibody and MPO have Female 73.3 45.2 0.0002 44% sequence homology, raising the possibility that cross- Ethnicity White 65 73.1 NS reactivity between TPO and MPO is responsible for the Afro-Caribbean 1.7 4.1 NS increased thyroid disease in patients with AAV. Although it Indo-Asian 26.7 14.2 0.02 has been reported that anti-TPO and anti-MPO antibodies can Mixed heritage 0 3.2 NS cross-react, this has not been proven in other studies, and it Unknown 6.7 6.8 NS may be that antibodies can only cross-react when peptide ANCA specificity 18,19,20 MPO 56.7 30.6 0.0004 sequences are denatured or reduced . Alternatively, PR3 33.3 54.3 0.006 rather than direct cross-reactivity, general loss of tolerance MPO and PR3 1.7 4.1 NS to peroxidases could explain this association. Negative 8.3 11 NS Our study has several limitations. This was a retrospective Current or ever smoker 70 73.5 NS study and there may have been missing data. It was difficult AAV: antineutrophil cytoplasmic antibody (ANCA)–associated vasculitis; to ascertain the temporal relationship between diagnosis of MPO: myeloperoxidase; PR3: proteinase 3; NS: not significant. thyroid disease and diagnosis of AAV. Although in some

Prendecki, et al: Thyroid disease in ANCA vasculitis 687

Downloaded on September 27, 2021 from www.jrheum.org Figure 1. Organ involvement of vasculitis in patients with AAV, with and without thyroid disease. *p < 0.05. AAV: antineutrophil cytoplasmic antibody–associated vasculitis; GI: gastrointestinal.

Table 2. Multivariate association of thyroid disease among patients with AAV. Values are n (%) unless otherwise specified.

Variables Thyroid Disease, No Thyroid Disease, OR (95% CI) p n = 60 n = 219

Female 44 (73.3) 99 (45.2) 3.3 (1.7–6.3) 0.0004 Renal disease 57 (95) 179 (81.7) 4.5 (1.3–15.4) 0.018 MPO-ANCA specificity 35 (58.3) 76 (34.7) 2.0 (1.1–3.8) 0.025 Indo-Asian ethnicity 16 (26.7) 31 (14.2) 1.5 (0.7–3.2) 0.246

AAV: antineutrophil cytoplasmic antibody (ANCA)–associated vasculitis; MPO: myeloperoxidase. patients it was clear that their hypothyroidism was secondary Given the high prevalence of thyroid disease, particularly to treated or burned-out hyperthyroidism, it was not possible hypothyroidism, in our group of patients with AAV, we would to be certain of the etiology of hypothyroidism in others. suggest that patients diagnosed with AAV should be assessed Another limitation is that information regarding the presence for evidence of thyroid disease or antithyroid antibodies of TPO antibodies was extremely limited. Of the patients through periodic testing of thyroid function. with thyroid disease and AAV, 20 (33.3%) had anti-TPO antibodies measured, with 6 patients having positive antibody ACKNOWLEDGMENT titers; 3 of these patients had MPO-ANCA and 3 had We acknowledge support from the National Institute for Health Research PR3-ANCA. It is possible that some of our patients with Imperial Biomedical Research Centre. negative anti-TPO antibodies had previously positive titers, although a study by Westman, et al suggested that thyroid REFERENCES antibodies persist over time despite patients being immuno- 1. Szyper-Kravitz M, Marai I, Shoenfeld Y. Coexistence of thyroid autoimmunity with other autoimmune diseases: friend or foe? suppressed and becoming ANCA-negative, suggesting that Additional aspects on the mosaic of autoimmunity. Autoimmunity the 14 patients with negative thyroid antibodies may have 2005;38:247-55. 21 been so throughout . Despite its limitations, our study is the 2. Biró E, Szekanecz Z, Czirják L, Danko K, Kiss E, Szabó NA, et al. largest cohort of patients with AAV reporting this association Association of systemic and thyroid autoimmune diseases. Clin with thyroid disease and MPO-ANCA, to our knowledge. Rheumatol 2006;25:240-5.

688 The Journal of Rheumatology 2018; 45:5; doi:10.3899/jrheum.170661

Downloaded on September 27, 2021 from www.jrheum.org 3. Pyne D, Isenberg DA. Autoimmune thyroid disease in systemic 13. Ishii R, Imaizumi M, Ide A, Sera N, Ueki I, Horie I, et al. A lupus erythematosus. Ann Rheum Dis 2002;61:70-2. long-term follow-up of serum myeloperoxidase antineutrophil 4. Perez B, Kraus A, Lopez G, Cifuentes M, Alarcon-Segovia D. cytoplasmic antibodies (MPO-ANCA) in patients with Graves Autoimmune thyroid disease in primary Sjogren’s syndrome. Am J disease treated with propylthiouracil. Endocr J 2010;57:73-9. Med 1995;99:480-4. 14. Vanderpump MP. The epidemiology of thyroid disease. Br Med Bull 5. Draibe J, Salama AD. Association of ANCA associated vasculitis 2011;99:39-51. and rheumatoid arthritis: a lesser recognized overlap syndrome. 15. Englund M, Merkel PA, Tomasson G, Segelmark M, Mohammad Springerplus 2015;4:50. AJ. Comorbidities in patients with antineutrophil cytoplasmic 6. Douglas G, Bird K, Flume P, Silver R, Bolster M. Wegener’s antibody-associated vasculitis versus the general population. granulomatosis in patients with rheumatoid arthritis. J Rheumatol J Rheumatol 2016;43:1553-8. 2003;30:2064-9. 16. Schultheiss UT, Daya N, Grams ME, Seufert J, Steffes M, Coresh J, 7. Derrett-Smith EC, Nihtyanova SI, Harvey J, Salama AD, Denton et al. Thyroid function, reduced kidney function and incident CP. Revisiting ANCA-associated vasculitis in systemic sclerosis: chronic kidney disease in a community-based population: the clinical, serological and immunogenetic factors. Rheumatology Atherosclerosis Risk in Communities study. Nephrol Dial 2013;52:1824-31. Transplant 2017;32:1874-81. 8. Levy JB, Hammad T, Coulthart A, Dougan T, Pusey CD. Clinical 17. Lo JC, Chertow GM, Go AS, Hsu CY. Increased prevalence of features and outcome of patients with both ANCA and anti-GBM subclinical and clinical hypothyroidism in persons with chronic antibodies. Kidney Int 2004;66:1535-40. kidney disease. Kidney Int 2005;67:1047-52. 9. Lionaki S, Hogan SL, Falk RJ, Joy MS, Chin H, Jennette CE, et al. 18. Haapala AM, Hyoty H, Parkkonen P, Mustonen J, Soppi E. Association between thyroid disease and its treatment with ANCA Antibody reactivity against thyroid peroxidase and myeloperoxidase small-vessel vasculitis: a case-control study. Nephrol Dial in autoimmune thyroiditis and systemic vasculitis. Scand J Immunol Transplant 2007;22:3508-15. 1997;46:78-85. 10. Tanaka A, Maeda K, Sawai K, Okuda J, Sugawara A, Kuwahara T. 19. Banga JP, Tomlinson RW, Doble N, Odell E, McGregor AM. Concealed hypothyroidism in patients with myeloperoxidase Thyroid microsomal/thyroid peroxidase autoantibodies show antineutrophili cytoplasmic autoantibodies- (MPO-ANCA) positive discrete patterns of cross-reactivity to myeloperoxidase, renal disease. Clinical Nephrol 1999;52:91-5. lactoperoxidase and horseradish peroxidase. Immunology 11. Harper L, Chin L, Daykin J, Allahabadia A, Heward J, Gough SC, et 1989;67:197-204. al. Propylthiouracil and carbimazole associated-antineutrophil 20. Freire BA, Paula ID, Paula F, Kallenberg CG, Limburg PC, Queluz cytoplasmic antibodies (ANCA) in patients with Graves’ disease. TT. Absence of cross-reactivity to myeloperoxidase of anti-thyroid Clin Endocrinol 2004;60:671-5. microsomal antibodies in patients with autoimmune thyroid 12. Noh JY, Asari T, Hamada N, Makino F, Ishikawa N, Abe Y, et al. diseases. Am J Med Sci 2001;321:109-12. Frequency of appearance of myeloperoxidase-antineutrophil 21. Westman KW, Bygren PG, Ericsson UB, Hoier-Madsen M, cytoplasmic antibody (MPO-ANCA) in Graves’ disease patients Wieslander J, Erfurth EM. Persistent high prevalence of thyroid treated with propylthiouracil and the relationship between antibodies after immunosuppressive therapy in subjects with MPO-ANCA and clinical manifestations. Clin Endocrinol glomerulonephritis. A prospective three-year follow-up study. Am J 2001;54:651-4. Nephrol 1998;18:274-9.

Prendecki, et al: Thyroid disease in ANCA vasculitis 689

Downloaded on September 27, 2021 from www.jrheum.org